Odds API Arbitrage Scanner

A Python repo for scanning bookmaker prices to find risk-free arbitrage bets,

Free key and account (Plug and play) Get a free Odds API key | Coverage | Docs | Pricing

What this does

• Loads upcoming sports events and bookmaker odds.
• Finds the best available price for each selection.
• Calculates implied probability and projected arbitrage margin.
• Builds a stake plan for equalized payout.
• Runs without credentials in mock mode.
• Supports live odds-api.net requests with ODDS_API_KEY.
• Can send top opportunities to a Discord webhook.

This project is a working starter, not betting advice. Prices move, markets suspend, limits apply, and execution can fail.

Quickstart

python -m venv .venv
.\.venv\Scripts\activate
python -m pip install -e ".[dev]"
python -m odds_api_arbitrage_scanner --mock --stake 100

Expected mock output:

1. City FC vs Rovers
   Market: h2h
   Profit: 7.88%
   Implied probability: 0.9270
   Best prices:
   - City FC: 2.58 at Bookmaker B
   - Draw: 3.62 at Bookmaker A
   - Rovers: 3.80 at Bookmaker A

Live API mode

Create a free key:

https://odds-api.net/signup

Then run:

$env:ODDS_API_KEY="your_api_key"
$env:ODDS_API_BASE_URL="https://api.odds-api.net/v1"
python -m odds_api_arbitrage_scanner --sport rugby-league --league NRL --market h2h --limit 20 --stake 100

For JSON output:

python -m odds_api_arbitrage_scanner --mock --format json

Discord alerts

$env:DISCORD_WEBHOOK_URL="https://discord.com/api/webhooks/..."
python -m odds_api_arbitrage_scanner --mock --stake 100

The alert includes the event, projected margin, best bookmaker for each selection, and stake allocation.

CLI options

--mock                 Use bundled mock odds.
--sport                Sport filter, for example rugby-league.
--league               League filter, for example NRL.
--market               Market to scan. Default: h2h.
--limit                Maximum events to load. Default: 25.
--stake                Total stake for allocation. Default: 100.
--min-profit           Minimum projected profit percent. Default: 0.
--format table|json    Output format.
--discord-webhook      Optional Discord webhook URL.

How the scanner works

For each event and market, the scanner:

1. Groups outcomes by selection.
2. Keeps the highest decimal odds for each selection.
3. Converts the best prices into implied probability.
4. Flags an opportunity when the implied probability is below 1.0.
5. Calculates stake sizing so each winning selection returns roughly the same payout.

Example formula:

implied_probability = (1 / best_home_odds) + (1 / best_away_odds)
profit_percent = ((1 / implied_probability) - 1) * 100

Project structure

.
├── src/odds_api_arbitrage_scanner/
│   ├── api.py
│   ├── cli.py
│   ├── config.py
│   ├── discord.py
│   ├── mock_data.py
│   └── scanner.py
├── tests/
├── docs/
├── examples/
├── .env.example
├── pyproject.toml
└── README.md

Testing

python -m pytest

Useful links

• Odds API homepage
• Create an API key
• API docs
• Coverage explorer
• Pricing

Safety

This code is for odds analysis and developer education. It does not place bets. Arbitrage is not risk-free. Odds can change between fetch and execution, accounts can be limited, markets can suspend, results can void, and availability varies by jurisdiction.

License

MIT